Study Finds No Causal Link Between Coffee Consumption, Arthritis Risk

There is no evidence that coffee intake causes an increased risk of developing rheumatoid arthritis (RA), according to a study published in Medicine, challenging earlier observational research that suggested a causal relationship between caffeine and arthritis risk.¹

To come to this conclusion, researchers analyzed genetic data from more than 97,000 individuals of European ancestry, including 22,350 cases of RA and 74,823 controls. Researchers used genome-wide association study (GWAS) data of 263,464 individuals of European ancestry to assess whether genetic predisposition to higher coffee intake was linked to RA, osteoarthritis (OA), or osteoporosis (OP).

Reconsidering Coffee’s Role in RA Risk

The relationship between coffee and RA has been inconsistent across prior studies. A 2014 meta-analysis suggested coffee might increase RA risk, potentially due to caffeine’s ability to promote pro-inflammatory immune responses in vitro.² However, other large cohort studies in the US and Denmark found no association between coffee intake and RA incidence.^3,4

By leveraging genetic data, the Mendelian randomization design helped avoid confounding factors such as smoking, diet, or recall bias.¹ According to the authors, this type of analysis has not been conducted to explore the relationship between coffee intake and bone health prior to this study.

“As the world’s most widely consumed beverage, coffee has biological effects that could potentially interact with autoimmune mechanisms,” the authors wrote. “However, current research findings regarding the association between coffee consumption and RA remain inconclusive.”

Genetic Instrument Selection and SNP Exclusions

Investigators used genetic variants associated with habitual coffee consumption as instrumental variables. Nineteen single-nucleotide polymorphisms (SNPs) met genome-wide significance thresholds (P < 5 × 10^–8) and had F-statistics above 10, ensuring strong instrument validity.

Because some of the variants were linked to potential confounders in the PhenoScanner database, they were excluded in sensitivity analyses. Specifically, 10 SNPs were identified as overlapping with RA and OA risk factors:

• rs574367
• rs10188334
• rs1877723
• rs74904971
• rs29647
• rs16903275
• rs2472297
• rs56094641
• rs476828
• rs56113850

After removing these variants and using inverse variance weighting (IVW) as the primary analytic method, the investigators determined that coffee consumption was not a causal factor for RA (OR, 0.84; 95% CI, 0.24-2.93).

Further refinement came from the MR-PRESSO analysis, which flagged 3 more SNPs for potential pleiotropy: rs1260326, rs73075167, and rs34060476. Excluding these also did not alter the null findings in IVW analysis (OR, 0.83; 95% CI, 0.14-4.78). Although the weighted median method suggested an elevated risk (OR, 1.30; 95% CI, 1.05-1.62), this estimate was inconsistent with the IVW model and not supported by heterogeneity testing, the authors noted.

“Our findings do not support a causal relationship between coffee intake and the risk of OA, RA, and OP,” they concluded.

Study limitations include that the analysis was restricted to individuals of European ancestry, limiting generalizability to other populations. Additionally, it did not capture coffee preparation methods or specific consumption levels, which may influence the concentrations of bioactive compounds.

References

1. Yang H, Yang L, Li Z, Liu H. Causality between coffee intake and bone health: A Mendelian randomization study. Medicine (Baltimore). 2025;104(37):e44533. doi:10.1097/MD.0000000000044533
2. Lee YH, Bae SC, Song GG. Coffee or tea consumption and the risk of rheumatoid arthritis: a meta-analysis. Clin Rheumatol. 2014;33(11):1575-1583. doi:10.1007/s10067-014-2631-1
3. Karlson EW, Mandl LA, Aweh GN, Grodstein F. Coffee consumption and risk of rheumatoid arthritis. Arthritis Rheum. 2003;48(11):3055-3060. doi:10.1002/art.11306
4. Pedersen M, Stripp C, Klarlund M, Olsen SF, Tjønneland AM, Frisch M. Diet and risk of rheumatoid arthritis in a prospective cohort. J Rheumatol. 2005;32(7):1249-1252.